N-oxyazinium salts are known to be photoinitiators for photocrosslinking and photopolymerization as described for example in U.S. Pat. Reissues 27,922 and 27,925 (both Heseltine et al.). Since most N-oxyazinium salt initiators absorb light in UV region of the electromagnetic spectrum, it is common practice to employ a photosensitizer co-initiator to increase their spectral response.
It is generally accepted that photosensitizing co-initiators function by absorption of a photon that results in excitation of an electron from an occupied molecular orbital to a higher energy, unoccupied orbital. The spin of the electron excited to the higher energy orbital corresponds to that which it exhibits in its original orbital or ground state. Thus, the photosensitizer in its initially formed excited state is in a singlet excited state. The lifetime of the singlet excited state is limited, typically less than a few nanoseconds. The excited photosensitizer can return from its singlet excited state directly to its original ground state, dissipating the captured photon energy. Alternatively, the singlet excited state photosensitizer in some instances undergoes intersystem crossing through spin inversion to another excited state, referred to as a triplet state, wherein lifetimes are typically in the microsecond to millisecond range. Since photosensitizer co-initiators that exhibit triplet states have longer lifetimes, the presence of the photosensitizer co-initiators provides a much longer time period for reaction.
GB Publication 2,083,832 (Specht et al.) describes photocurable compositions that comprise N-oxyazinium salts and co-initiators based on amino-substituted ketocoumarin triplet photosensitizers. The amino-substituted ketocoumarins exhibit very high intersystem crossing (or triplet state generation) efficiencies ranging well above 10%. U.S. Pat. No. 4,743,528 (Farid et al.) disclose a photocurable composition comprising an N-oxyazinium salt, an N-oxyazinium activator, and a photosensitizer having a reduction potential that in relation to the reduction potential of the N-oxyazinium salt activator is at most 0.1 V more positive, and an electron rich amino-substituted benzene. Similarly, U.S. Pat. No. 4,743,530 (Farid et al.) describes photocurable compositions containing an N-oxyazinium salt activator and a dye based photosensitizer with maximum absorption above 550 nm and having a reduction potential relative to that of N-oxyazinium salt activator is at most 0.1 V more positive.
N-oxyazinium salts have been demonstrated as useful sources of radicals for photoinitiating polymerization. Single electron transfer from an excited electron donor (D*) to an N-oxyazinium salt results in N—O bond cleavage and the formation of an oxy radical, as shown below in Equation (1).

Although a number of dye-based, as well as, triplet ketocoumarin-based photosensitizing co-initiators have been used to initiate photopolymerization using N-oxyazinium salts, most of them have limited curing speed. This is usually due to overall lower quantum efficiency of the process. The quantum yield of a radiation-induced process is the number of times that a defined event occurs per photon absorbed by the system. The event could be the decomposition of a reactant molecule.
In the case of photopolymerization using N-oxyazinium salts and ketocoumarin triplet photosensitizers, the overall quantum efficiency of oxy radical generation is less than or equal to the triplet formation efficiency (the limiting quantum efficiency being defined as state efficiency for reaction times the quantum yield for formation of the reacting state). With dye-based photosensitizers, the overall quantum efficiency is expected to be even lower due to a shorter lifetime of excited dye.
To increase the overall efficiency of a photocuring process, some degree of amplification is necessary. That is, amplification of photoreactions where one photon leads to the transformation of several reactant molecules to products. In some cases, the commercial viability of certain systems can depend on whether a relatively modest amplification, for example, in the 10 to 100 times range, could be achieved. This depends usually upon limitations on exposure time, light intensity, or a combination that can be imposed on a specific use.
In most known amplified photochemical processes, amplification is based on photochemical generation of a species that is subsequently used to catalyze another reaction. Very few examples of amplified photoreactions are known where one photon leads to the transformation of several reactant molecules to products. Most of these quantum-amplified electron-transfer processes involve radical cation reactions, such as valence isomerization, for example, the transformation of hexamethyldewarbenzene to hexamethylbenzene, or the cyclization or cycloreversion between two olefin moieties and a cyclobutane, where quantum yields as high as several hundred have been obtained in polar solvents. In these reactions, the chain is propagated via electron transfer from a reactant molecule (R) to the radical cation of the product (P.+).
Another type of chain-amplified photoreaction involves two reactants where one is oxidized (leading, for example, to dehydrogenation) and the other is reduced. A different kind of chain reaction involving two reactants is that of onium salts. In these reactions, upon one electron reduction an onium salt (Ar—X+) undergoes fragmentation to yield an aryl radical, which in turn takes a hydrogen atom from an alcohol to give an α-hydroxyl radical. Chain propagation occurs through electron transfer from the α-hydroxyl radical to another onium salt molecule.
Amplified photosensitized electron transfer reactions of N-methoxypyridinium salts with alcohols of diverse structures were recently demonstrated (Shukla et al., J. Org. Chem. 70, 6809-6819.) to achieve quantum efficiencies of ˜10-20, even at modest reactant concentrations of 0.02-0.04 M, and in spite of the endothermicity of the critical electron transfer step from the intermediate α-hydroxy radical to the pyridinium salt. These reactions can be initiated by a number of singlet or triplet sensitizers, with varying degrees of initiation efficiencies that can be as high as 2.
A number of photoinitiators and photocurable compositions have been developed and commercialized to carry out free radical chain polymerization. In most of these methods, free radicals are produced by either of two pathways:
(1) the photoinitiator undergoes excitation by energy absorption with subsequent decomposition into one or more radicals, or
(2) the photoinitiator undergoes excitation and the excited species interacts with a second compound (by either energy transfer or a redox reaction) to form free radicals from the latter or former compound(s).
Most known photoinitiators have only moderate quantum yields (generally less than 0.4), indicating that the conversion of light radiation to radical formation needs to be made more efficient. Thus, there are continuing opportunities for improvements in the use of photoinitiators in free radical polymerization.
In photopolymerization technology, there still exists a need for highly amplified photochemistry, and easy to prepare and easy to use photocurable compositions. The need for amplified photocurable compositions is particularly acute where absorption of light by the reaction medium may limit the amount of energy available for absorption by the photoinitiators. For example, in the preparation of color filter resists, highly pigmented resists are required for high color quality. With the increase in pigment content, the curing of color resists becomes more difficult. The same is true for the UV-photocurable inks, for example offset printing inks, which also are loaded with pigments. Hence, there is a need for a photocurable composition having a higher sensitivity and excellent resolution properties. In addition, there is a need for such photocurable compositions to meet the industrial properties such as high solubility, thermal stability, and storage stability.
Besides the challenges above that are often encountered in free radical curing, there is an additional challenge of free radical photocuring inhibition by the presence of oxygen. Oxygen inhibition has always been a problem for photocuring of compositions containing multifunctional acrylate monomers or oligomers using a photoinitiated radical polymerization (for example, see Decker et al., Macromolecules 18 (1985) 1241.). This oxygen inhibition is due to the rapid reaction of carbon centered propagating radicals with oxygen molecules to yield peroxyl radicals. These peroxyl radicals are not as reactive towards carbon-carbon unsaturated double bonds and therefore do not initiate or participate in any photopolymerization reaction. Oxygen inhibition usually leads to premature chain termination that results in incomplete photocuring. Thus, many photocuring processes must be carried out in inert environments (for example, under nitrogen or argon), making such processes more expensive and difficult to use in industrial and laboratory settings.
Various methods have been proposed to overcome oxygen inhibition of photocuring:
(1) Amines that can undergo a rapid peroxidation reaction can be added to consume the dissolved oxygen. However, the presence of amines in acrylate-containing compositions can cause yellowing in the resulting photocured composition, create undesirable odors, and soften the cured composition because of chain transfer reactions. Moreover, the hydroperoxides thus formed will have a detrimental effect on the weathering resistance of the UV-cured composition.
(2) Dissolved oxygen can be converted into its excited singlet state by means of a red light irradiation in the presence of a dye sensitizer. The resulting 1O2 radical will be rapidly scavenged by a 1,3-diphenylisobenzofuran molecule to generate a compound (1,2-dibenzoylbenzene) that can work as a photoinitiator (Decker, Makromol. Chem. 180 (1979), p. 2027). However, the photocured composition can become colored, in spite of the photobleaching of the dye, prohibiting this technique for use in various products.
(3) The photoinitiator concentration can be increased to shorten the UV exposure during which atmospheric oxygen diffuses into the cured composition. This technique can also be used in combination with higher radiation intensities. Oxygen inhibition can further be reduced by using high intensity flashes that generate large concentrations of initiator radicals reacting with oxygen, but hydroperoxides are also formed.
(4) Free radical photopolymerization can be carried out under inert conditions (Wight, J. Polym. Sci.: Polym. Lett. Ed. 16 (1978) 121), which is the most efficient way to overcome oxygen inhibition. Nitrogen is typically continuously used to flush the photopolymerizable composition during UV exposure. On an industrial UV-curing line, which cannot be made completely airtight, nitrogen losses can be significant, thus making the process expensive and inefficient. This is an even greater concern if argon is used to provide an inert environment.
Other less common ways of overcoming oxygen inhibition of acrylate photopolymerization include using a wax barrier and performing UV exposure under water. Each of these techniques has disadvantages that have made them less likely for commercial application.
Thus, there is a need to provide highly efficient photocuring or photopolymerization of acrylate-containing compositions using N-oxyazinium salts without the need for inert environment or use of other known methods for reducing oxygen inhibition of free radical formation and reaction.